Social capital impact on individual health due to neighbours or the neighbourhood?

Mohnen, S.M., Groenewegen, P.P., Völker, B., Flap, H., Subramanian, S.V. Social capital impact on individual health due to neighbours or the neighbourhood? European Journal of Public Health: 2010, 20(suppl. 1), p. 42. Abstract. 3rd European Public Health Conference 'Integrated Public Health', 10-13 November 2010, Amsterdam.
Background: We test two different perspectives on social capital and its association with health; a macro perspective focusing on resources on the neighbourhood level, and a micro perspective focusing on the individual level personal network. So far, most researchers inquired into these perspectives separately, and studies combining and testing both views are scarce. More specifically, we examine whether both macro and micro social capital are associated with individual’s health, whether a shortage of one type of social capital can be compensated by the other, and whether both types of social capital accumulate (that people who are rich in both types of social capital are better off). Methods: The ‘Dutch Housing Demand Survey’ is used and combined with relevant macro information (Statistics Netherlands). Results: In a logistic multilevel analysis we found that both forms of social capital were positively associated with self-perceived health. Having only few contacts to friends or family members is compensated by a high level of neighbourhood social capital. Micro and macro social capital also accumulate. We provide two possible explanations: micro-social capital might be an access to macro social capital, or a high social capital neighbourhood increases the likelihood of frequent contact to friends (e.g. the neighbourhood has a good image and friends like to come by). Conclusions: Macro social capital is associated with self-perceived health and it matters even while controlling for micro social capital. We propose that future research should take the interrelatedness of the two types of social capital into account.